Biochimie (1992) 74, 581-584 © Soci6t6 fran~aise de biochimie et biologie mol6culaire / Elsevier, Paris

581

Short communication

Overproduction and purification of lysyl-tRNA synthetase encoded by the herC gene of E coli Y Nakamura, K Kawakami Department of Tumor Biology, The Institute of Medical Science, The University of Tokyo, 4-6-1 Shirokanedai, Minato-ku, Tokyo 108, Japan (Received 20 January i 992; accepted 17 February 1992)

Summary - - Lysyl-tRNA synthetases are synthesized from two distinct genes in E coli, lysS and lysU, but neither gene product has

been purified distinctively by using overproducing systems. The lysS gene has been identified by a her(? mutation which restores maintenance of the mutant ColE1 replicon. The herC gene product was overproduced by using a tac promoter fusion and purified to homogeneity. The purified HerC protein possesses a lysyl-tRNA synthetase activity as predicted by the sequence identity of herC to lysS. The procedure is useful for rapid mass-scale purification of lysyl-tRNA synthetase.

lysyl-tRNA synthetase I herC I overproduction Introduction E coli has two forms of lysyl-tRNA synthetase (LysRS). The major form of LysRS is encoded by lysS located at 62 min on the E coli chromosome [ 1] and the minor form is encoded by iysU located at 93.5 rain [2]. This is a rare exception, as far as is known, to the rule o f one synthetase per amino acid. lysS is in the same operon as the prfB gene which encodes peptide-chain-release factor 2 [3], and is expressed constitutively under all measured growth conditions. On the other hand, lysU is normally silent and expressed under selected conditions, including growth at high temperatures and growth in the presence of certain nutrients or metabolites [4, 5]. Neither the mechanism of differential regulation nor the physiological significance of these two genes is presently understood. The herC mutation has been defined in a host mutant which restored maintenance o f a replicationdefective CoiEI plasmid carrying a primer RNA mutation [6]. The herC gene has been cloned and sequenced [3]. The primary protein sequence is similar to the yeast cytoplasmic lysyl-tRNA synthetase [7]. L6v~que et al have cloned the lysS gene by DNA hybridization using the oligonucleotide probe Abbreviations: LysRS, lysyl-tRNA synthetase; bp, base pair(s); IPTG, isopropyl-l-thio,~-D-galactoside; SDS, sodium dodecyl sulfate.

synthesized from the NH.,-terminal sequence of LysRS [8]. The deduced lysS sequence coincides with that of herC. In this article, we constructed the herCoverexpressing plasmid, developed a rapid purification procedure of the gene product, and established biochemically that the purified herC protein carries lysyl-tRNA synthetase activity (this work was presented at the NATO/EEC workshop on 'Post-Transcriptional Control of Gene Expression' held in Goslar, Germany, April 1990 [9]).

Materials and methods Bacteria and plasmids

Bacterial strains (E coli K-12) were C600 (F- thr leu tonA lac thi supE44) [10] and AD20 (thi tsx::Tn5 lonlO0 rpoHl6 zhfi:TnlO A(pro-lac)[F'[laciq lacZAMl5 lacY+ pro+l) kindly provided by T Yura (Kyoto University, Japan). Plasmids pKK945 and pKK948 were described previously [6]. The HerC-overproducing plasmid pKK990 was constructed by inserting the herC-coding sequence (1660-bp HindIII-Nrul fragment isolated from pKK948) at the Hind|II site of the pTTQ8 plasmid [11], under control of the tac promoter, by using the HindIII linker. LysRS assay

The activity of the synthetase was determined in crude sonic extracts or fractions of column chromatography. Protein was determined by using the Bio-Rad protein assay reagent (Richmond, CA). The assay was run at 37°C for 2 rain essentially as described previously [12] in a solution (100 ~tl) con-

582 taining 0.1 M Tris-HCI (pH 7.6), 1 mM ATP, 10 mM MgCI,, 0.1 mg of crude E coli B tRNA or 4 I.tg of E coli lysyl-tRNA (kindly provided by S Yokoyama, University of Tokyo, Japan), 1.5 laM L-[l'~2llysine (0.05 I.tCi, Amersham, UK), and 1 lag of crude extract or 2 I11 of fraction after appropriate dilution if necessary. Acid insoluble incorporation was monitored after filtration and washing through GF/C filters (Whatman BioSystems Ltd, UK).

hydroxyapatite column were pooled and proteins were precipitated by adding solid ammonium sulfate to 73.2% saturation. After centrifugation, the precipitate was dissolved in ! ml of 0.1 M PEG, and loaded onto a column (1 x 60 cm) of Seph~cryl S-300 Superfine (Pharmacia Fine Chemicals, Sweden) equilibrated with 0.1 M PEG buffer. The fractions with the LysRS activity were pooled and stored at -80°C after concentration by using a DEAE-cellulose mini-column as described above.

LysRS purification Enzyme purification was carried out at 0--4°C, unless otherwise stated, according to the method presented by /~,kesson and Lundvik [ 13 I after several modifications. Overproducing strain AD20 cells carrying pKK990 were grown in 1 I of L broth containing 50 lag ampicillin per ml at 32°C and 1 mM IPTG was added at the mid-log phase. After 90 rain, the cells were harvested and used for the enzyme purification. Throughout the purification, each fraction was analyzed by SDS-polyacrylamide gel electrophoresis followed by staining with Coomassie brilliant blue to detect oversynthesized HerC and also assayed for LysRS activity. The harvested cells (3.6 g cell paste) were ground with glass beads (75-150 lain) and suspended in 50 ml solution containing 0.02 M Tris-HCl, pH 7.6, 4 mM EDTA and 10% glycerol. After centrifugation for 30 rain at 15 000 g, the extract was precipitated with 0.2 col of a 5% streptomycin sulfate solution added dropwise with continuous stirring. After 30 min of stirring the precipitate was removed by centrifugation for 30 rain at 15 000 g. To the supernatant solution (53 ml) was gradually added 13.3 g of solid ammonium sulfate (41.1% saturation). After stirring for 30 rain, the precipitate was removed by centrifugation for 30 min at 15 000 g and solid ammonium sulfate (7.23 g) was slowly added to the supernatant (58 ml) (62.9% saturation). After 30 rain stirring, the precipitate was collected by centrifugation for 30 rain at 15 000 g, and dissolved in 10 ml solution containing 2 mM EDTA and 10% glycerol (EG solution). Conductivity of the sample solution was adjusted to that of 0.1 M potassium phosphate buffer containing 2 mM EDTA and 10% glycerol (PEG buffer). The sample was applied to a column of DEAE--cellulose (DE52; Whatman BioSystems Ltd, UK) (1.4 × 17 cm column size) equilibrated with 0.1 M PEG buffer, Elution was achieved by using a linear gradient of 0.1--0.45 M PEG buffer (60 ml). The fractions containing HerC were pooled, diluted with EG solution until the conductivity reached 0.05 M PEG, and directly applied to a column of hydro~tyapatite (Bio-Gel HTP; Bio-Rad, Richmond, CA) equilibrated with 0.05 M PEG. Elution was achieved by using a linear gradient of 0.05-0.15 M PEG buffer (60 ml). The fractions containing HerC were pooled and concentrated by 0.05/0.45 M stepwise chromatography using a mini-scale column of DEAE-cellulose. The purified HerC was dialyzed against 0. ! M potassium phosphate buffer (pH 7.5) containing 2 mM EDTA, 50% glycerol and 0.5 mM dithiothreitol, and stored at -80°C. Non-overproducing strain C600 cells carrying the pKK945 plasmid were cultured in 20 1 of L broth containing 15 lag chloramphenicol per ml at 37°C. The harvested cell paste (35 g) was used to purify the LysRS enzyme. The purification proce~lures are essentially the same as described above except that the buffer and column volumes were scaled up five times and protein solutions were dialyzed to adjust conductivity, instead of the dilution method, before column chromatography. The LysRS fractions eluted from a

The other method NHz-terminal sequence of LysRS was determined by an automatic protein sequencer (Applied Biosystems, CA).

Results and discussion E coli cells transformed with plasmid pKK945 (nonoverproducer) carrying herC yielded = three-fold increase in the LysRS activity in the crude sonic extracts in vitro (data not shown). We first attempted to purify the LysRS e n z y m e from pKK945-bearing C600 cell paste (35 g) according to the procedure described by Akesson and L u n d v i k [13] with several modifications. Cell extracts were fractionated with 41.1-62.9% saturated a m m o n i u m sulfate and subjected to chromatography using a DEAE-cellulose column. When the peak fractions containing LysRS were pooled and applied to a hydroxyapatite column, the LysRS activity split in two major peaks, ie LysRSI and LysRS-II (fig 1). These two forms do not correspond to the constitutive and thermo-inducible LysRS enzymes because the NH2-terminal sequences o f these enzymes determined by an automatic protein sequencer after complete purification coincide exactly with each other; ie S E Q H A Q G A , which is the NH,terminal sequence of the lysS enzyme but not o f the lysU e n z y m e [8, 14]. The NH2-terminal methionine was excised in both forms. LysRS-I and LysR$-II fractions were further purified individually by gel filtration and compared with each other. W h e n the purified enzymes were reloaded on a hydroxyapatite column, they were eluted exactly at the same positions as before. We failed to detect any associated protein or any difference in the activity, protein size or isoelectric point of these two forms (data not shown). These results might be explained by assuming that an unidentified protein modification (or processing, though less likely) m a y be the cause of two forms o f LysRS. Final yields of enzymes were 0.5 and 0.7 m g o f LysRS-I and LysRS-II, respectively. Plasmid pKK990 (overproducer) was constructed by placing the herC segment under control o f the tac promoter in p T r Q 8 . C600 cells transformed with p K K 9 9 0 yielded an increased activity of LysRS in the crude-extract assay upon induction with I F F G , but less than six-fold. The overproduced HerC protein

583 2

~]¢.NK

A

0

0.45M iS ......y........."......... :s

1

t B

__~ 0 E

F R A C T I O N NO.

o.a C

x

W

'

~

'°:°!

Gio | ..... . . . . _-" -_ 2"0 ~

o

r,, !!

o-o,,,." 1"0

../..

/0 so

40

]~

0.1SM

"

.'-"...................."'°-

•

]ooo,,j

30

, 3 O 1M I

F R A C T I O N NO.

,1_.1 =°/ 20

LysRS'I ........ LysRS'II

"'I1,,," ao

.....o.,p..M... ~: ~- I D

° °

1

~'1

,40

500

F R A C T I O N NO.

1

1(}

,.,,,,,,., tl ::// 20

% 3"0

. iO

5"0

60 0

F R A C T I O N NO.

Fig 1. Purification of LysRS. (A, B) C600 cells carrying the pKK945 plasmid were used for purifying LysRS as described in Materials and methods. A. DEAE-cellulose column chromatography. B. Hydroxyapatite column chromatography. C, D. AD20 cells carrying pKK990 were grown and harvested after addition of IPTG as described in Materials and methods. The cell paste was used for purifying the HerC protein by monitoring the overproduced protein and the LysRS activity. C. DEAE-cellulose column chromatography. D. Hydroxyapatite column chromatography. Note that the salt gradients (presented by dotted lines) used are slightly different between A and C, or B and D.

degraded rapidly under C600 strain background (data not shown). Therefore, we used the strain AD20 (lonlO0 rpoH16) defective in the Ion and heat-inducible proteases to accumulate the HerC protein. As shown in figures 2A and B, AD20 cells transformed with pKK990 overproduced and markedly accumulated HerC at 32°C upon addition of IPTG for 90 min. The resulting cell paste (3.6 g) was used to purify overproduced HerC by ammonium sulfate fractionation and column chromatography using DEAEcellulose and hydroxyapatite. The fractions containing overproduced HerC were monitored by Coomassie staining after SDS-polyacrylamide gel electrophoresis. It is evident that the first DEAE-cellulose column chromatography yields more than 80% pure HerC

fractions (fig 2C). Again, HerC split in two peaks via hydroxyapatite-column chromatography (fig 1). The minor and major peaks correspond to LysRS-I and LysRS-II, respectively, as judged from the elution point. It seems likely that LysRS-II represents a primary form because hyperproduction of HerC gave rise to more LysRS-II fractions. Yet no difference was observed in the enzyme features of the two forms isolated from overproducing (pKK990) and honorerproducing (pKK945) cells. Final yields of LysRS-I and LysRS-II were 0.35 and 2.7 rag, respectively. Throughout the purification, overproduced HerC was exactly copurified with the LysRS activity. These data are interpreted as biochemical evidence for identity of HerC to LysRS. The procedure described in this

584

(ARe)12 13 14 •

(C)

(D)

15 16 17 18 19 20 21 22 23 24

7

m

m

m

Iiwo~

~

,--HerC

Fig 2. Coomassie-staining analysis of overproduced and purified HerC protein after SDS-polyacrylamide gel electrophoresis. Total proteins of AD20{pKK990] cells prior to the IPTG induction (A). Total proteins of AD20[pKK990] cells after a 90-min IPTG induction (B). DEAE-cellulose fractions from figure 2C (C). Purified HerC protein (D). article is useful for rapid mass-scale purification of lysyl-tRNA synthetase.

Acknowledgments We thank S Yokoyama for E coli lysyl-tRNA and BS Powell for reading the manuscript. This work was supported in part by grants from The Ministry of Education, Science and Culture, Japan,

References 1 Emmerich RV, Hirshfield IN(1987) Mapping of the constitutive lysyl-tRNA synthetase gene of Escherichia coli K-12. JBacteriol 169, 5311-5313

2 Van Bogelen RA, Vaughn V, Neidhardt FC (1983) Gene for heat-inducible lysyl-tRNA synthetase (lysU) maps near carla in Escherichia coli. J Bacteriol 153, 1066-1068 3 Kawakami K, J6nsson YH, Bj&k GR, Ikeda H, Nakamura Y (1988) Chromosomal location and structure of the operon encoding peptide-chain-release factor 2 of Escherichia coli. Proc Natl Acad Sci USA 85, 5620-5624 4 Hirshfield IN, Bloch PL, VanBogelen RA, Neidhardt FC (1981) Multiple forms of lysyl-transfer ribonucleic acid synthetase in Escherichia coli. J Bacteriol ! 46, 345-351 5 Hirshfield IN, Yeh FM, Sawyer LE (1975) Metabolites influence control of lysine transfer ribonucleic acid synthetase formation in Escherichia coli K-12. Proc Natl Acad Sci USA 72, 1364-1367 6 Kawakami K, Naito S, Inoue N, Nakamura Y, Ikeda H, Uchida H (1989) Isolation and characterization of herC, a mutation of Escherichia coli affecting maintenance of ColEI. Mol Gen Genet 219, 333-340 7 Gampel A, Tzagoloff A (1989) Homology of aspartyl- and lysyl-tRNA synthetases. Proc Natl Acad Sci USA 86, 6023-6027 8 L6v~que F, Plateau P, Dessen P, Blanquet S (1990) Homology of lysS and lysU, the two Escherichia coli genes encoding distinct lysysl-tRNA synthetase species. Nucleic Acids Res 18,305-312 9 Nakamura Y, Kawakami K, Mikuni O (1990) Alternative translation and functional diversity of release factor 2 and lysyl-tRNA synthetase. In: Post-Transcriptional Control of Gene Expression (McCarthy ~EG, Tuite MF, eds) Springer-Verlag Press, Berlin, 455--464 10 Appleyard RK (1954) Segregation of lambda lysogenicity during bacterial recombination in E coil K-12. Genetics 39, 429-439 11 Stark M~R (1987) Multicopy expression vectors carrying the lac repressor gene for regulated high-level expression of genes in Escherichia coli. Gene 51,255-267 12 Hirshfield IN, Tenreim R, VanBogelen RA, Neidhardt FC (1984) Escherichia coli K-12 lysyl-tRNA synthetase mutant with a novel reversion pattern. Y Bacteriol 158, 615-620 13 Akesson B, Lundvik L (1978) Simultaneous purification and some properties of aspartate:tRNA ligase and seven other amino-acid:tRNA ligases from Escherichia coil. Fur J Biochem 83, 29-36 14 Clark RL, Neidhardt FC (1990) Roles of the two lysyltRNA synthetases of Escherichia coli: analysis of nueleotide sequences and mutant behavior. J Bacteriol 172, 3237-3243